Metraipolis › archive › dc_514cebf63326

POD-RoPE: Rotary Position Encoding over Physical Observation Distributions for Embodied AI — arXiv:2610.33745

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.33745
License (as archived)
CC0-1.0
Archived
2026-10-27T08:00:26.409Z
SHA-256 of the archived text
01b6754353890083a9db924f98f7f99a1c91ab4e3bce07347922334922509f4b
Stable URL of this page
https://metraipolis.com/documento/dc_514cebf63326

How to cite

POD-RoPE: Rotary Position Encoding over Physical Observation Distributions for Embodied AI — arXiv:2610.33745 — arXiv (en), https://arxiv.org/abs/2610.33745 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_514cebf63326 (Metraipolis archive dc_514cebf63326, sha256 01b6754353890083a9db924f98f7f99a1c91ab4e3bce07347922334922509f4b)
BibTeX
@misc{dc514cebf63326,
  title = {POD-RoPE: Rotary Position Encoding over Physical Observation Distributions for Embodied AI — arXiv:2610.33745},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.33745},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_514cebf63326; sha256 01b6754353890083a9db924f98f7f99a1c91ab4e3bce07347922334922509f4b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33745. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Embodied perception requires positional representations that capture both physical geometry and the spatial and temporal extent of observations. We introduce POD-RoPE (Physical Observation Distribution RoPE), which represents token observation footprints as distributions over camera centers, viewing directions, and sampling times. Under independent sampling, expected rotations encode relative relationships between distributions; when each distribution collapses to a single address, the construction recovers pointwise RoPE. The same rotation statistics support Q/K matching and coherence-dependent V/O transformations, allowing observation geometry to guide both information selection and content aggregation. Per-token pre-aggregation avoids increasing sequence length or processing pairs of observation nodes. We evaluate POD-RoPE across four robot policy architectures, simulation and real-world manipulation, and complementary reconstruction, depth-estimation, and novel-view-synthesis tasks. Integrating POD-RoPE into Fast-WAM improves mean success from 91.83% to 92.88% across all 50 RoboTwin 2.0 tasks, and from 32.5% to 52.5% on two real-robot tasks on Sharpa North. Controlled reconstruction ablations support the contributions of distribution encoding and V/O transformations. In H20 model-block benchmarks, POD-RoPE achieves median forward-plus-backward speedups of $1.91\times$ over P-RoPE and $5.34\times$ over RayRoPE, with a median latency overhead of 17.4% over Standard RoPE. These results support physical observation distributions as an effective and computationally efficient basis for geometry-aware attention.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.33745 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-27T07:55:30.846Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).